| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "global_step": 3504, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002574486301369863, | |
| "loss": 3.1277, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 2.02591609954834, | |
| "eval_runtime": 179.7935, | |
| "eval_samples_per_second": 19.467, | |
| "eval_steps_per_second": 2.436, | |
| "eval_wer": 0.9999398659009592, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.0002146404109589041, | |
| "loss": 1.4111, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 1.125086784362793, | |
| "eval_runtime": 179.8421, | |
| "eval_samples_per_second": 19.462, | |
| "eval_steps_per_second": 2.435, | |
| "eval_wer": 0.889383324814336, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.00017183219178082193, | |
| "loss": 0.8461, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_loss": 0.8204647302627563, | |
| "eval_runtime": 183.3614, | |
| "eval_samples_per_second": 19.088, | |
| "eval_steps_per_second": 2.389, | |
| "eval_wer": 0.724375357046213, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.00012902397260273972, | |
| "loss": 0.5042, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "eval_loss": 0.6116402745246887, | |
| "eval_runtime": 180.1904, | |
| "eval_samples_per_second": 19.424, | |
| "eval_steps_per_second": 2.431, | |
| "eval_wer": 0.5462882227367029, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 8.621575342465752e-05, | |
| "loss": 0.3072, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "eval_loss": 0.5506712198257446, | |
| "eval_runtime": 180.8043, | |
| "eval_samples_per_second": 19.358, | |
| "eval_steps_per_second": 2.423, | |
| "eval_wer": 0.45055473706365196, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 4.340753424657534e-05, | |
| "loss": 0.2181, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.5213309526443481, | |
| "eval_runtime": 179.0608, | |
| "eval_samples_per_second": 19.546, | |
| "eval_steps_per_second": 2.446, | |
| "eval_wer": 0.4177215189873418, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 5.993150684931506e-07, | |
| "loss": 0.1608, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_loss": 0.5160985589027405, | |
| "eval_runtime": 181.598, | |
| "eval_samples_per_second": 19.273, | |
| "eval_steps_per_second": 2.412, | |
| "eval_wer": 0.40187618389007485, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 3504, | |
| "total_flos": 7.417092095436165e+18, | |
| "train_loss": 0.9386324367770865, | |
| "train_runtime": 4216.4813, | |
| "train_samples_per_second": 13.281, | |
| "train_steps_per_second": 0.831 | |
| } | |
| ], | |
| "max_steps": 3504, | |
| "num_train_epochs": 8, | |
| "total_flos": 7.417092095436165e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |